Unmanned aerial vehicle radar early warning system and method thereof

Abstract

The invention provides an unmanned aerial vehicle radar early warning system and method, and the method comprises the steps: carrying out the feature extraction of a radar echo signal based on a radar system collection airspace, and obtaining the time domain feature and frequency domain feature of each signal; signal screening is carried out based on feature screening conditions in combination with the time domain features and the frequency domain features, and suspected unmanned aerial vehicle target signals are determined; performing signal association on continuous multiple frames of suspected unmanned aerial vehicle target signals, determining a spatial position difference value and a signal feature similarity of the suspected unmanned aerial vehicle target signals in adjacent frames, and determining a target signal sequence based on the spatial position difference value and the signal feature similarity; determining the movement speed and the movement direction of the suspected unmanned aerial vehicle target signal in the continuous frames based on the target signal sequence, and determining whether the movement of the suspected unmanned aerial vehicle target has continuity and regularity based on the movement speed and the movement direction; if yes, it is judged that the unmanned aerial vehicle target exists, and early warning information is generated and output. The reliability of radar early warning of the unmanned aerial vehicle is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle, and particularly relates to an unmanned aerial vehicle radar early warning system and a method thereof. BACKGROUND

[0002] With the rapid development of unmanned aerial vehicle technology, unmanned aerial vehicles are increasingly widely used in civilian and commercial fields, and the problems of unmanned aerial vehicle "black flight" and "wrong flight" have posed a serious threat to airspace safety, important facility protection and public safety. To cope with these threats, unmanned aerial vehicle radar early warning methods have emerged as the times require, and the core goal is to detect, identify and track unmanned aerial vehicles through radar systems and timely issue early warning information so as to take measures such as interception and driving away. In the prior art, a common unmanned aerial vehicle radar early warning method is a signal detection and early warning mechanism based on a fixed threshold, that is, a fixed radar echo signal strength threshold is set, and when the detected signal strength exceeds the threshold, it is determined that there is an unmanned aerial vehicle target and early warning is issued.

[0003] However, affected by the flight height, flight attitude, distance from the radar station and complex electromagnetic environment (such as ground clutter and electromagnetic interference) of the unmanned aerial vehicle, the radar echo signal strength of the unmanned aerial vehicle will fluctuate greatly. When the unmanned aerial vehicle is flying at a long distance and low altitude or encounters strong electromagnetic interference, the echo signal strength may be lower than the fixed threshold, resulting in the radar system failing to detect the target and causing a missed alarm; and when there is a strong clutter or interference signal, the signal strength may exceed the fixed threshold, causing the system to misjudge as an unmanned aerial vehicle target and resulting in a false alarm. The problems of missed alarm and false alarm seriously reduce the reliability of unmanned aerial vehicle radar early warning, and it is difficult to meet the high requirements for early warning accuracy and timeliness in actual applications. SUMMARY

[0004] The present application provides an unmanned aerial vehicle radar early warning system and method to reduce missed alarms and false alarms and improve the reliability of unmanned aerial vehicle radar early warning.

[0005] In a first aspect, the present application provides an unmanned aerial vehicle radar early warning method, comprising:

[0006] Based on the radar system collecting the radar echo signals of the airspace, the radar echo signals are subjected to feature extraction to obtain the time domain features and frequency domain features of each signal in the radar echo signals;

[0007] Based on the feature screening conditions combined with the time domain features and frequency domain features of each signal, signal screening is performed to determine the suspected unmanned aerial vehicle target signals in the radar echo signals;

[0008] signal correlation is performed on the suspected UAV target signals of continuous multiple frames, spatial position difference and signal feature similarity of the suspected UAV target signals in adjacent frames are determined, and target signal sequence is determined based on the spatial position difference and the signal feature similarity;

[0009] Based on the target signal sequence, the motion speed and the motion direction of the suspected UAV target signal in continuous frames are determined, and whether the motion of the suspected UAV target has persistence and regularity is determined based on the motion speed and the motion direction.

[0010] If it has persistence and regularity, it is determined that there is a UAV target, a warning information is generated and output.

[0011] According to the UAV radar warning method provided by the application, the determination of the motion speed and the motion direction of the suspected UAV target signal in continuous frames based on the target signal sequence comprises:

[0012] Based on the time interval and the change amount of the spatial position between the adjacent frame signals in the target signal sequence, the instantaneous speed components of the adjacent frame signals on the three-dimensional coordinate axes in the spatial rectangular coordinate system are determined.

[0013] Based on the instantaneous speed components of the adjacent frame signals, the instantaneous resultant speed of the adjacent frame signals and the motion direction angle component of the adjacent frame signals in the spatial rectangular coordinate system are determined.

[0014] The instantaneous resultant speeds of the adjacent frame signals are arranged in sequence according to the frame sequence to generate the motion speed of the suspected UAV target signal in continuous frames, and the motion direction angle components of the adjacent frame signals in the spatial rectangular coordinate system are combined to generate the motion direction of the suspected UAV target signal in continuous frames.

[0015] According to the UAV radar warning method provided by the application, the determination of whether the motion of the suspected UAV target has persistence and regularity based on the motion speed and the motion direction comprises:

[0016] Based on the motion speed of the continuous frames, the change rate of the instantaneous resultant speed between the continuous frames is determined, and based on the motion direction of the continuous frames, the direction change amount between the continuous frames is determined.

[0017] Based on the change rate of the instantaneous resultant speed between the continuous frames, a speed persistence index between the continuous frames is determined, and based on the direction change amount between the continuous frames, a direction persistence index between the continuous frames is determined.

[0018] Based on the motion speed of the continuous frames and the average speed in the target signal sequence, a speed regularity deviation amount between the continuous frames is determined, and based on the motion direction of the continuous frames and the relative direction in the target signal sequence, a direction regularity deviation amount between the continuous frames is determined.

[0019] If the speed persistence index between the continuous frames is less than or equal to the preset speed index threshold value, and the direction persistence index between the continuous frames is less than or equal to the preset direction index threshold value, it is determined that the motion has persistence;

[0020] If the speed regularity deviation amount between the continuous frames is less than or equal to the speed deviation threshold value, and the direction regularity deviation amount between the continuous frames is less than or equal to the preset direction deviation threshold value, it is determined that the motion has regularity.

[0021] According to the unmanned aerial vehicle radar early warning method provided by the application, the signal correlation of the continuous multiple frames of suspected unmanned aerial vehicle target signals is performed to determine the spatial position difference and signal feature similarity of the suspected unmanned aerial vehicle target signals in adjacent frames, which comprises:

[0022] A space rectangular coordinate system is established with the phase center of the radar antenna as the origin; the horizontal axis of the space rectangular coordinate system points to the front of the radar, the vertical axis is horizontally to the right, and the vertical axis is vertically upward;

[0023] Based on the spatial position parameters and signal feature parameters of each frame of suspected unmanned aerial vehicle target signals in the space rectangular coordinate system, a feature vector is constructed; the signal feature parameters include time domain features and frequency domain features;

[0024] For the kth frame of suspected unmanned aerial vehicle target signal, the position difference base of the suspected unmanned aerial vehicle target signal in the kth frame is determined according to the first spatial position parameter in the feature vector combined with the second spatial position parameter in the feature vector of all suspected unmanned aerial vehicle target signals in the k+1th frame;

[0025] According to the first signal feature parameter in the feature vector combined with the second signal feature parameter in the feature vector of all suspected unmanned aerial vehicle target signals in the k+1th frame, the feature similarity base of the suspected unmanned aerial vehicle target signal in the kth frame is determined;

[0026] Based on the position difference base and the feature similarity base of the suspected unmanned aerial vehicle target signal in the kth frame, the spatial position difference and the signal feature similarity of the suspected unmanned aerial vehicle target signal in the adjacent frame are determined.

[0027] According to the unmanned aerial vehicle radar early warning method provided by the application, the spatial position difference and the signal feature similarity of the suspected unmanned aerial vehicle target signal in the adjacent frame are determined based on the position difference base and the feature similarity base of the suspected unmanned aerial vehicle target signal in the kth frame, which comprises:

[0028] Based on the position difference base of the suspected unmanned aerial vehicle target signal in the kth frame, the first distribution feature coefficient of all suspected unmanned aerial vehicle target signals in the kth frame in the k+1th frame is determined;

[0029] determine a second distribution characteristic coefficient of all suspected unmanned aerial vehicle target signals in the k+1th frame based on the characteristic similarity basis of the suspected unmanned aerial vehicle target signals in the kth frame;

[0030] determine a spatial position difference value of the suspected unmanned aerial vehicle target signals in the adjacent frames based on the position difference basis of the suspected unmanned aerial vehicle target signals in the kth frame and the first distribution characteristic coefficient;

[0031] determine a signal characteristic similarity of the suspected unmanned aerial vehicle target signals in the adjacent frames based on the characteristic similarity basis of the suspected unmanned aerial vehicle target signals in the kth frame and the second distribution characteristic coefficient.

[0032] According to the unmanned aerial vehicle radar early warning method provided by the application, the target signal sequence is determined based on the spatial position difference value and the signal characteristic similarity, comprising:

[0033] construct a correlation degree matrix based on the spatial position difference value and the signal characteristic similarity of the suspected unmanned aerial vehicle target signals in the adjacent frames; the matrix elements in the correlation degree matrix represent the correlation degree between the suspected unmanned aerial vehicle target signals in the kth frame and the k+1th frame;

[0034] determine a correlation signal pair between the kth frame and the k+1th frame based on the correlation degree matrix satisfying a preset correlation condition; each correlation signal pair comprises a first signal and a second signal;

[0035] determine a cooperative coefficient of each correlation signal pair based on the first spatial position difference value and the first signal characteristic similarity of the first signal and the second spatial position difference value and the second signal characteristic similarity of the second signal;

[0036] determine a target signal sequence based on the cooperative coefficient of each correlation signal pair between the kth frame and the k+1th frame.

[0037] According to the unmanned aerial vehicle radar early warning method provided by the application, the target signal sequence is determined based on the cooperative coefficient of each correlation signal pair between the kth frame and the k+1th frame, comprising:

[0038] combine the target correlation signal pairs satisfying a preset cooperative condition based on the cooperative coefficient of each correlation signal pair between the kth frame and the k+1th frame, to obtain an initial signal sequence segment between the kth frame and the k+1th frame;

[0039] determine a spatial continuity index of each initial signal sequence segment based on the spatial position difference value of each signal in the target correlation signal pair;

[0040] determine a characteristic consistency index of each initial signal sequence segment based on the signal characteristic similarity of each signal in the target correlation signal pair;

[0041] The candidate signal sequence fragments are spliced across frames to obtain the target signal sequence.

[0042] In a second aspect, the present application further provides a UAV radar early warning system, which is applied to the UAV radar early warning method as described in the first aspect, and comprises:

[0043] A signal collection and extraction module is configured to collect radar echo signals of a space domain based on a radar system, and extract features of the radar echo signals to obtain time domain features and frequency domain features of each signal in the radar echo signals.

[0044] A signal screening module is configured to screen signals based on feature screening conditions in combination with the time domain features and the frequency domain features of each signal to determine suspected UAV target signals in the radar echo signals.

[0045] A signal correlation module is configured to correlate the suspected UAV target signals in consecutive multiple frames to determine spatial position difference values and signal feature similarities of the suspected UAV target signals in adjacent frames, and determine a target signal sequence based on the spatial position difference values and the signal feature similarities.

[0046] A motion trajectory analysis module is configured to determine motion speed and motion direction of the suspected UAV target signals in consecutive frames based on the target signal sequence, and determine whether the motion of the suspected UAV target signals has continuity and regularity based on the motion speed and the motion direction.

[0047] An early warning determination and output module is configured to determine that there is a UAV target if the motion has continuity and regularity, generate early warning information, and output the early warning information.

[0048] The present application further provides an electronic device, which comprises a memory configured to store a computer software program, and a processor configured to read and execute the computer software program to realize the UAV radar early warning method as described above.

[0049] The present application further provides a non-transitory computer readable storage medium, which stores a computer software program, and the computer software program is executed by a processor to realize the UAV radar early warning method as described above.

[0050] The present application further provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to realize the UAV radar early warning method as described above.

[0051] The unmanned aerial vehicle radar early warning method provided by the embodiment of the present application avoids missing the target due to a single signal strength index by extracting multiple features to screen suspected unmanned aerial vehicle target signals, and further confirms the authenticity of the suspected unmanned aerial vehicle target signals through multi-frame signal correlation and motion trajectory analysis, so that the unmanned aerial vehicle radar early warning method can detect the suspected unmanned aerial vehicle target signals as long as the features and the motion conform to the rules even if the signal strength fluctuates, thereby reducing the missed alarms. In addition, the multi-frame correlation requires that the spatial positions and the features are similar, which excludes instantaneous interference signals, and the motion trajectory analysis judges the continuity and regularity of the motion, which excludes irregular interference signals, thereby reducing the false alarms and improving the reliability of the unmanned aerial vehicle radar early warning. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 FIG. 1 is a structural schematic diagram of the unmanned aerial vehicle radar early warning method provided by the present application;

[0053] Figure 2 FIG. 2 is a flow schematic diagram of the unmanned aerial vehicle radar early warning system provided by the present application;

[0054] Figure 3 FIG. 3 is an embodiment schematic diagram of the electronic device provided by the present application. DETAILED DESCRIPTION

[0055] The technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0056] In the description of the present application, the terms "first", "second" are used only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0057] In the description of the present application, the term "for example" is used to indicate "as an example, illustration or explanation". Any embodiment described as "for example" in the present application is not necessarily interpreted as more preferred or more advantageous than other embodiments. The following description is given in order to enable any person skilled in the art to implement and use the present application. In the following description, details are listed for the purpose of explanation. It should be understood that those skilled in the art can realize the present application without using these specific details.

[0058] In other instances, well-known structures and processes have not been described in detail in order to avoid unnecessarily obscuring the description of the present application. Accordingly, the present application is not intended to be limited to the described embodiments, but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0059] Optionally, referring to Figure 1 , Figure 1 is a structural schematic diagram of the unmanned aerial vehicle radar early warning method provided by the present application, and the unmanned aerial vehicle radar early warning method comprises:

[0060] Step 10, based on the radar system, collecting the radar echo signal of the airspace, and extracting the features of the radar echo signal to obtain the time domain feature and the frequency domain feature of each signal in the radar echo signal.

[0061] Optionally, the unmanned aerial vehicle early warning system continuously scans the specified airspace through the radar system equipped therewith to collect the radar echo signal in the airspace. The radar system emits electromagnetic waves, and when the electromagnetic waves encounter objects (including possible unmanned aerial vehicle targets, birds, fixed obstacles, etc.) in the airspace, reflections are generated to form radar echo signals, which are received by the unmanned aerial vehicle early warning system.

[0062] Further, the unmanned aerial vehicle early warning system extracts features from the collected radar echo signal. Time domain feature extraction mainly analyzes the characteristics of the signal in the time dimension, including peak amplitude (the maximum amplitude value of the signal on the time axis), pulse width (the time length of the signal duration) and pulse repetition period (the time interval between adjacent two pulse signals). Frequency domain feature extraction is to convert the time domain signal to the frequency domain through Fourier transform and other methods, analyze the frequency characteristics of the signal, including center frequency (the main frequency where the signal energy is concentrated) and bandwidth (the frequency range occupied by the signal).

[0063] In an embodiment, the radar system is scanning a certain surrounding airspace, and when the electromagnetic wave encounters a flying unmanned aerial vehicle, the electromagnetic wave is reflected by the unmanned aerial vehicle to form a radar echo signal. After receiving this signal, in the time domain feature extraction, the unmanned aerial vehicle early warning system analyzes the radar echo signal and finds that the peak amplitude is 5V (i.e. the maximum amplitude value of the radar echo signal on the time axis is 5V); by measuring the time from the beginning to the end of the signal, the pulse width is determined to be 200 microseconds; by recording the start time difference of adjacent two pulse signals, the pulse repetition period is obtained to be 10 milliseconds.

[0064] In the frequency domain feature extraction, the UAV early warning system performs Fourier transform on the radar echo signal, and after conversion to the frequency domain, it is found that the signal energy is mainly concentrated near 3 GHz, so the center frequency is 3 GHz; at the same time, the signal extends from 2.9 GHz to 3.1 GHz in frequency, so the bandwidth is 0.2 GHz.

[0065] Step 20, based on the feature screening condition, the time domain features and the frequency domain features of each signal are combined to screen the signals, and the suspected UAV target signal in the radar echo signal is determined.

[0066] Further, the feature screening condition is set in advance, and the feature screening condition is based on the typical signal features of the UAV target. For example, the radar echo signal of the UAV target usually has a peak amplitude, a pulse width, a pulse repetition period, a center frequency and a bandwidth within a certain range.

[0067] Therefore, the UAV early warning system compares the time domain features and the frequency domain features of each radar echo signal with the feature screening condition one by one. When the features of a certain radar echo signal are all within the range specified by the feature screening condition, the signal is determined as a suspected UAV target signal.

[0068] Continue with the above example, the set feature screening condition is as follows: the peak amplitude is between 3V-8V, the pulse width is between 100 microseconds-500 microseconds, the pulse repetition period is between 5 milliseconds-20 milliseconds, the center frequency is between 2GHz-4GHz, and the bandwidth is between 0.1GHz-0.5GHz.

[0069] The time domain features of a certain radar echo signal extracted in step 10 are: peak amplitude 5V, pulse width 200 microseconds, pulse repetition period 10 milliseconds; the frequency domain features are: center frequency 3GHz, bandwidth 0.2GHz. The UAV early warning system compares these features with the screening condition, and finds that all the features are within the specified range, so it determines that the radar echo signal is a suspected UAV target signal.

[0070] In addition, there is a radar echo signal with a peak amplitude of 2V, which is lower than the 3V in the screening condition, and the UAV early warning system determines that the signal does not meet the condition and does not list it as a suspected UAV target signal.

[0071] Step 30, signal correlation is performed on the suspected UAV target signals of consecutive multiple frames to determine the spatial position difference and signal feature similarity of the suspected UAV target signals in adjacent frames, and based on the spatial position difference and the signal feature similarity, the target signal sequence is determined.

[0072] Further, the UAV early warning system performs signal correlation on the suspected UAV target signals in continuous multiple frames, i.e. analyzes the relationship of the suspected UAV target signals in adjacent frames. For two adjacent frames, the spatial positions of each suspected UAV target signal in the two frames are determined, and the spatial position difference (i.e. the distance difference of the two signals in the spatial coordinates) between them is calculated. At the same time, the signal feature similarity of the suspected UAV target signals in adjacent frames is calculated, where the signal features include time domain features and frequency domain features. The signal feature similarity is obtained by calculating the similarity of each feature, specifically as the process of steps 301 to 305.

[0073] Further, the UAV early warning system determines, according to the spatial position difference and the signal feature similarity of the suspected UAV target signals in adjacent frames, that the two signals belong to the same target in different frames, determines the target signal sequence, specifically as the process of steps 306 to 309.

[0074] Step 40, based on the target signal sequence, the motion speed and the motion direction of the suspected UAV target signals in continuous frames are determined, and based on the motion speed and the motion direction, it is determined whether the motion of the suspected UAV target has continuity and regularity.

[0075] Further, the UAV early warning system calculates the motion speed of each suspected UAV target signal in the target signal sequence according to the spatial position and the time information of the suspected UAV target signals in continuous frames. The motion speed is calculated by dividing the spatial position difference of the signals in adjacent frames by the time interval between the two frames. At the same time, the UAV early warning system determines the motion direction of the suspected UAV target signal according to the change direction of the spatial position, such as horizontal east, vertical up, etc., specifically as the process of steps 401 to 403.

[0076] Further, the UAV early warning system analyzes the change of the motion speed and the motion direction, and judges whether the motion of the suspected UAV target has continuity (i.e. the motion in continuous frames is not interrupted, and the speed and the direction do not suddenly disappear) and regularity (i.e. the change of the motion speed is within a reasonable range, and the motion direction remains stable or changes regularly, such as uniform linear motion, uniform circular motion, etc.), specifically as the process of steps 404 to 407.

[0077] Step 50, if it has continuity and regularity, it is determined that there is a UAV target, and the early warning information is generated and output.

[0078] Further, when it is determined that the motion of the suspected UAV target has continuity and regularity, the UAV early warning system determines that there is a UAV target.

[0079] Further, the UAV early warning system generates early warning information, which contains the current position, motion speed, motion direction, discovery time and other key information of the UAV target.

[0080] In an embodiment, it is determined that the motion of a suspected UAV target is persistent and regular, and thus the UAV early warning system determines that there is a UAV target.

[0081] The early warning information generated by the UAV early warning system is: "On July 19, 2025, at 10:00, a UAV target was discovered at airspace coordinates (110, 200, 50), with a motion speed of 5 meters per second and a motion direction of horizontal along the positive x-axis direction. Please pay attention to monitoring by relevant personnel." Then, the UAV early warning system displays the early warning information through its equipped display screen, and triggers the alarm to emit a warning sound, to ensure that relevant personnel can timely know the situation.

[0082] The embodiment of the present application avoids missing the target due to a single signal strength indicator by extracting multiple features for suspected UAV target signal screening, and further confirms the authenticity of the suspected UAV target signal through multi-frame signal correlation and motion trajectory analysis. Even if the signal strength fluctuates, as long as the characteristics and motion conform to the rules, it can be detected, thereby reducing false alarms. Furthermore, through multi-frame correlation, the spatial position and characteristics are required to be similar, which excludes transient interference signals. Motion trajectory analysis judges the persistence and regularity of the motion, which excludes irregular interference signals, thereby reducing false alarms and improving the reliability of the UAV radar early warning.

[0083] In an embodiment, the process of steps 301 to 305 includes:

[0084] Step 301, a spatial rectangular coordinate system is established with the phase center of the radar antenna as the origin. The horizontal axis of the spatial rectangular coordinate system points to the front of the radar, the vertical axis is horizontally to the right, and the vertical axis is vertically upward.

[0085] Optionally, the UAV early warning system constructs a spatial rectangular coordinate system with its equipped radar antenna phase center as the origin. Among them, the horizontal axis (usually set as X axis) points to the front of the radar, that is, the main emission direction of the radar electromagnetic wave; the vertical axis (usually set as Y axis) is horizontally to the right, which is perpendicular to the horizontal axis in the same horizontal plane; the vertical axis (usually set as Z axis) is vertically upward, which together with the horizontal axis and the vertical axis forms a right-hand screw rule.

[0086] In an embodiment, the phase center of the radar antenna is located at a fixed position, and a space rectangular coordinate system is established with this position as the origin (0, 0, 0). The front of the radar is the direction of the airport runway, so the X-axis points to the direction in which the airport runway extends; the Y-axis points horizontally to the right and to the empty land on the right side of the runway; and the Z-axis points vertically upward and to the sky. When a suspected unmanned aerial vehicle target signal appears in the radar monitoring range, its position in the coordinate system can be converted into the (x, y, z) coordinates through the radar measurement data, for example, (100 meters, 50 meters, 30 meters), indicating that the suspected unmanned aerial vehicle target signal is 100 meters in front of the radar, 50 meters to the right, and 30 meters high.

[0087] Step 302: Based on the spatial position parameters of each suspected unmanned aerial vehicle target signal in each frame in the space rectangular coordinate system and the signal characteristic parameters thereof, a feature vector is constructed.

[0088] Further, for each suspected unmanned aerial vehicle target signal in each frame, the unmanned aerial vehicle early warning system extracts the spatial position parameters (i.e., the X-axis coordinate, the Y-axis coordinate, and the Z-axis coordinate) in the space rectangular coordinate system and the signal characteristic parameters (time domain features: peak amplitude, pulse width, and pulse repetition period; frequency domain features: center frequency and bandwidth), and combines the spatial position parameters in the space rectangular coordinate system and the signal characteristic parameters in a fixed order to form a feature vector of the suspected unmanned aerial vehicle target signal.

[0089] In an embodiment, there is a suspected unmanned aerial vehicle target signal in the kth frame, and its spatial position parameters in the space rectangular coordinate system are (150 meters, 60 meters, 40 meters). In the signal characteristic parameters, the time domain features are a peak amplitude of 4.5V, a pulse width of 180 microseconds, and a pulse repetition period of 8 milliseconds, and the frequency domain features are a center frequency of 2.8GHz and a bandwidth of 0.15GHz. The parameters are combined in the order of “X-axis coordinate, Y-axis coordinate, Z-axis coordinate, peak amplitude, pulse width, pulse repetition period, center frequency, and bandwidth” to construct a feature vector of [150 meters, 60 meters, 40 meters, 4.5V, 180 microseconds, 8 milliseconds, 2.8GHz, 0.15GHz].

[0090] Step 303: For the suspected unmanned aerial vehicle target signal in the kth frame, the first spatial position parameter in the feature vector thereof is combined with the second spatial position parameter in the feature vectors of all suspected unmanned aerial vehicle target signals in the k+1th frame to determine the position difference base of the suspected unmanned aerial vehicle target signal in the kth frame.

[0091] Further, the UAV early warning system extracts the first spatial position parameter (i.e. the X-axis coordinate, Y-axis coordinate, Z-axis coordinate of the signal) in the feature vector of the suspected UAV target signal in the kth frame. Meanwhile, the second spatial position parameter (X-axis coordinate, Y-axis coordinate, Z-axis coordinate of each signal) in the feature vector of all suspected UAV target signals in the k+1th frame is extracted.

[0092] Further, the UAV early warning system calculates the absolute value of the coordinate difference of the signal in the kth frame and each signal in the k+1th frame on the three coordinate axes, and then calculates the weighted sum according to the preset weight (usually the same weight for the three coordinate axes) to obtain the position difference base.

[0093] In an embodiment, the first spatial position parameter of a suspected UAV target signal in the kth frame is (200 meters, 80 meters, 50 meters). There are two suspected UAV target signals in the k+1th frame, and their second spatial position parameters are (205, 82, 51) and (180, 70, 45) respectively. For the first signal in the k+1th frame, the absolute value of the X-axis coordinate difference is |200-205| = 5 meters, the Y-axis is |80-82| = 2 meters, and the Z-axis is |50-51| = 1 meter, and the weight is 1 / 3. The position difference base is (5+2+1)*(1 / 3) = 8 / 3 ≈ 2.67 meters.

[0094] For the second signal in the k+1th frame, the absolute value of the X-axis coordinate difference is |200-180| = 20 meters, the Y-axis is |80-70| = 10 meters, and the Z-axis is |50-45| = 5 meters. The position difference base is (20+10+5)*(1 / 3) = 35 / 3 ≈ 11.67 meters.

[0095] Step 304, according to the first signal feature parameter in the feature vector and the second signal feature parameter in the feature vector of all suspected UAV target signals in the k+1th frame, the feature similarity base of the suspected UAV target signal in the kth frame is determined.

[0096] Further, the UAV early warning system extracts the first signal feature parameter (peak amplitude, pulse width, pulse repetition period, center frequency, bandwidth) in the feature vector of a suspected UAV target signal in the kth frame, and the second signal feature parameter (the same as the above five features) in the feature vector of all suspected UAV target signals in the k+1th frame. For each feature, the similarity (such as peak amplitude similarity = (smaller value / larger value)*100%) is calculated, and then the weighted average is calculated according to the preset weight (such as the same weight for each feature) to obtain the feature similarity base.

[0097] In an embodiment, the first signal characteristic parameter of a certain suspected unmanned aerial vehicle target signal in the kth frame is: peak amplitude 5V, pulse width 200 microseconds, pulse repetition period 10 milliseconds, center frequency 3GHz, and bandwidth 0.2GHz. The second signal characteristic parameter of a certain suspected unmanned aerial vehicle target signal in the k+1th frame is: peak amplitude 5.2V, pulse width 190 microseconds, pulse repetition period 10.1 milliseconds, center frequency 3.05GHz, and bandwidth 0.19GHz. The characteristic similarity degrees are calculated:

[0098] The peak amplitude similarity degree is (5 / 5.2)*100% ≈ 96.15%.

[0099] The pulse width similarity degree is (190 / 200)*100% = 95%.

[0100] The pulse repetition period similarity degree is (10 / 10.1)*100% ≈ 99.01%.

[0101] The center frequency similarity degree is (3 / 3.05)*100% ≈ 98.36%.

[0102] The bandwidth similarity degree is (0.19 / 0.2)*100% = 95%.

[0103] Each weight is 1 / 5, and the characteristic similarity degree base is (96.15% + 95% + 99.01% + 98.36% + 95%)*(1 / 5) = (483.52%)*(1 / 5) = 96.704%.

[0104] In step 305, based on the position difference base and the characteristic similarity degree base of the suspected unmanned aerial vehicle target signal in the kth frame, the spatial position difference and the signal characteristic similarity degree of the suspected unmanned aerial vehicle target signal in the adjacent frame are determined.

[0105] Further, the unmanned aerial vehicle early warning system determines the spatial position difference and the signal characteristic similarity degree of the suspected unmanned aerial vehicle target signal in the adjacent frame according to the position difference base and the characteristic similarity degree base of the suspected unmanned aerial vehicle target signal in the kth frame, specifically as the processes of steps 3051 to 3054.

[0106] The embodiment of the application realizes position quantization by establishing a unified spatial coordinate system, integrates multi-dimensional information by constructing a characteristic vector, provides a correlation basis by calculating a position difference base and a characteristic similarity degree base, and finally determines a spatial position difference and a signal characteristic similarity degree, so that the signals belonging to the same unmanned aerial vehicle target in the adjacent frames can be accurately identified, and the non-associated signals (such as signals generated by flying birds and clutter) can be excluded, thereby laying a solid foundation for subsequent formation of a target signal sequence and ensuring that the subsequent analysis of the motion characteristics of the unmanned aerial vehicle target is based on continuous and accurate signal data, and the target recognition accuracy and reliability of the unmanned aerial vehicle early warning are improved.

[0107] In one embodiment, steps 3051 to 3054 include:

[0108] Step 3051: Based on the position difference base of the suspected UAV target signal in the k-th frame, determine the first distribution characteristic coefficient of all suspected UAV target signals in the k+1-th frame.

[0109] Optionally, the drone early warning system collects the position difference base number of all suspected drone target signals in the k-th frame. The position difference base number is the basic data related to the position difference calculated between each suspected drone target signal in the k-th frame and all suspected drone target signals in the (k+1)-th frame.

[0110] Furthermore, the UAV early warning system calculates the integral of the probability density function of these position difference bases over a preset interval, and combines it with the standard deviation of the bases within that interval to determine the first distribution characteristic coefficient. The first distribution characteristic coefficient can reflect the overall distribution density and dispersion of the position difference bases.

[0111] In one embodiment, there are three suspected UAV target signals in the k-th frame: signal A, signal B, and signal C. The position difference base values ​​calculated for each signal and all suspected UAV target signals in the (k+1)-th frame are as follows: the position difference base values ​​for signal A are 2.67 meters, 8.33 meters, and 15.00 meters; for signal B, they are 3.33 meters, 9.00 meters, and 16.67 meters; and for signal C, they are 4.00 meters, 9.67 meters, and 17.33 meters. The UAV early warning system first determines the range of position difference base values ​​to be [0, 20] meters. It then calculates the probability density function f(x) for all position difference base values ​​within this range, where x represents the position difference base value. Integrating f(x) over the [0, 20] range yields an integral result I1 = 0.95. Finally, it calculates the standard deviation σ1 = 5.2 meters for all position difference base values ​​within this range.

[0112] First distribution characteristic coefficient a k The calculation formula is: a k =I1*exp(-(σ1) 2 / 2*(D0) 2 )

[0113] Wherein, D0 is the preset location distribution benchmark parameter, with a value of 10 meters.

[0114] Substitute the numerical value into the formula: a k The first distribution characteristic coefficient of all suspected UAV target signals in frame k is approximately 0.83, meaning that the first distribution characteristic coefficient of all suspected UAV target signals in frame k+1 is 0.83.

[0115] Step 3052, based on the feature similarity basis of the suspected UAV target signal in the kth frame, determine the second distribution feature coefficient of all suspected UAV target signals in the k+1th frame.

[0116] Further, the UAV early warning system collects the feature similarity basis of all suspected UAV target signals in the kth frame, and the feature similarity basis is the feature similarity related basic data calculated by each suspected UAV target signal in the kth frame and all suspected UAV target signals in the k+1th frame.

[0117] Further, the UAV early warning system determines the second distribution feature coefficient by calculating the function value of the cumulative distribution function of these feature similarity bases at the preset threshold, combined with the mean value of the feature similarity basis above the preset threshold, the second distribution feature coefficient can reflect the overall distribution concentration trend and the level of high similarity part of the feature similarity basis.

[0118] Continue the above embodiment, the feature similarity basis calculated by signal A, signal B, signal C in the kth frame and all suspected UAV target signals in the k+1th frame is as follows: the feature similarity basis of signal A is 96.70%, 82.50%, 65.30%; the feature similarity basis of signal B is 95.80%, 81.20%, 64.10%; the feature similarity basis of signal C is 94.90%, 80.30%, 63.50%. The preset threshold of the feature similarity basis is 60%. Calculate the function value F(60%) of the cumulative distribution function F(y) of the feature similarity basis at y=60%, F(60%)=1 (i.e. all bases are above 60%). Then calculate the mean value μ2=81.50% of all feature similarity bases above the preset threshold 60%. The calculation formula of the second distribution feature coefficient β k is as follows:

[0119] β k =F(60%)*(1-exp(-μ2 / μ0)).

[0120] Wherein, μ0 is the preset feature similarity reference parameter, and the value is 100%.

[0121] Substitute the numerical value into the formula: β k =0.558, that is, the second distribution feature coefficient of all suspected UAV target signals in the kth frame in the k+1th frame is 0.558.

[0122] Step 3053, based on the position difference basis of the suspected UAV target signal in the kth frame and the first distribution feature coefficient, determine the spatial position difference of the suspected UAV target signal in the adjacent frame.

[0123] Further, the UAV early warning system fuses and calculates the position difference base number of a suspected UAV target signal in the kth frame and the first distribution characteristic coefficient through a pre-designed calculation function, wherein the pre-designed calculation function can comprehensively consider the position difference base data of a single signal and the overall position distribution characteristics, so that the calculated spatial position difference not only reflects the actual position difference between two signals, but also integrates the influence of the overall distribution, and is more consistent with the position correlation characteristics of the UAV target signal in the actual scene.

[0124] Continuing the above embodiment, the position difference base number of signal A in the kth frame and a suspected UAV target signal in the k+1th frame is 2.67 meters, and the first distribution characteristic coefficient a k = 0.83. The calculation formula of the spatial position difference d k,k+1 is: d k,k+1 = (x k,k+1 *(1-a k )) / (1+exp(-x k,k+1 / (a k *L0))).

[0125] Wherein, x k,k+1 is the position difference base number of a suspected UAV target signal in the kth frame and a suspected UAV target signal in the k+1th frame, and L0 is a pre-set position reference length, which is 10 meters.

[0126] Substitute the numerical value into the formula: d k,k+1 ≈ 0.263 meters, that is, the spatial position difference of signal A in the kth frame and the suspected UAV target signal in the k+1th frame is 0.263 meters.

[0127] Step 3054, based on the feature similarity base number of the suspected UAV target signal in the kth frame and the second distribution characteristic coefficient, the signal feature similarity of the suspected UAV target signal in the adjacent frame is determined.

[0128] Further, the UAV early warning system fuses and calculates the feature similarity base number of a suspected UAV target signal in the kth frame and the second distribution characteristic coefficient through a pre-designed calculation function, wherein the pre-designed calculation function can comprehensively consider the feature similarity base data of a single signal and the overall feature distribution characteristics, so that the calculated signal feature similarity not only reflects the feature similarity degree between two signals, but also combines the overall feature distribution, and can more accurately reflect the feature correlation characteristics of the UAV target signal.

[0129] Continuing the above embodiment, the feature similarity base number of signal A in the kth frame and a suspected UAV target signal in the k+1th frame is 96.70%, and the second distribution characteristic coefficient β k = 0.558. The calculation formula of the signal feature similarity s k,k+1 is: sk,k+1 = (y k,k+1 * β k ) * (1 + tan (y k,k+1 * β k / S0) ).

[0130] Wherein, y k,k+1 is the feature similarity base number of a suspected unmanned aerial vehicle target signal in the kth frame and a suspected unmanned aerial vehicle target signal in the k+1th frame, and S0 is a preset similarity reference value, and the value is 1.

[0131] The numerical value is substituted into the formula s k,k+1 ≈ 0.81, that is, the signal feature similarity of the signal A in the kth frame and the suspected unmanned aerial vehicle target signal in the k+1th frame is 0.81.

[0132] The embodiment of the application calculates the first distribution characteristic coefficient and the second distribution characteristic coefficient, respectively quantifies the distribution characteristics of the position difference base number and the feature similarity base number from the overall distribution, and then calculates the spatial position difference and the signal feature similarity of the suspected unmanned aerial vehicle target signal in the adjacent frames by the distribution characteristics and the basic data of the single signal, so that the difference of the single signal and the overall distribution can be considered comprehensively, the calculated spatial position difference and signal feature similarity are more accurate and reasonable, and the reliability of whether the suspected unmanned aerial vehicle target signals in the adjacent frames belong to the same target is improved.

[0133] In an embodiment, the process of steps 306 to 309 comprises:

[0134] Step 306, based on the spatial position difference and the signal feature similarity of the suspected unmanned aerial vehicle target signals in the adjacent frames, a correlation matrix is constructed. The matrix elements in the correlation matrix represent the correlation degree between the suspected unmanned aerial vehicle target signals in the kth frame and the k+1th frame.

[0135] Optionally, the unmanned aerial vehicle early warning system constructs a correlation matrix for reflecting the correlation degree between signals by the spatial position difference and the signal feature similarity of the suspected unmanned aerial vehicle target signals in the adjacent frames, therefore, the number of suspected unmanned aerial vehicle target signals in the kth frame and the k+1th frame is determined, assuming that there are m suspected unmanned aerial vehicle target signals in the kth frame and n suspected unmanned aerial vehicle target signals in the k+1th frame, then the dimension of the correlation matrix is m*n. For each element in the correlation matrix, the value is determined by the correlation degree of the spatial position difference and the signal feature similarity of a suspected unmanned aerial vehicle target signal in the kth frame and a suspected unmanned aerial vehicle target signal in the k+1th frame. The embodiment of the application adopts a fusion calculation mode of the inverse ratio of the signal feature similarity and the spatial position difference to obtain the correlation degree value of the element, and the higher the correlation degree value, the greater the possibility that the two signals belong to the same unmanned aerial vehicle target.

[0136] In an embodiment, there are two suspected unmanned aerial vehicle target signals in the kth frame, signal A and signal B; there are three suspected unmanned aerial vehicle target signals in the k+1th frame, signal C, signal D and signal E. After calculation, the spatial position difference and signal feature similarity between each signal are as follows:

[0137] Signal A and signal C: the spatial position difference is 0.263 meters, and the signal feature similarity is 0.81.

[0138] Signal A and signal D: the spatial position difference is 0.85 meters, and the signal feature similarity is 0.65.

[0139] Signal A and signal E: the spatial position difference is 1.52 meters, and the signal feature similarity is 0.42.

[0140] Signal B and signal C: the spatial position difference is 0.31 meters, and the signal feature similarity is 0.78.

[0141] Signal B and signal D: the spatial position difference is 0.72 meters, and the signal feature similarity is 0.83.

[0142] Signal B and signal E: the spatial position difference is 1.35 meters, and the signal feature similarity is 0.56.

[0143] The unmanned aerial vehicle early warning system adopts the formula: correlation degree = signal feature similarity * (1 / space position difference) * 0.1 to calculate the correlation degree matrix element value, so the correlation degree between each signal is:

[0144] The correlation degree between signal A and signal C = 0.81 * (1 / 0.263) * 0.1 ≈ 0.81 * 3.802 * 0.1 ≈ 0.308.

[0145] The correlation degree between signal A and signal D = 0.65 * (1 / 0.85) * 0.1 ≈ 0.65 * 1.176 * 0.1 ≈ 0.076.

[0146] The correlation degree between signal A and signal E = 0.42 * (1 / 1.52) * 0.1 ≈ 0.42 * 0.658 * 0.1 ≈ 0.028.

[0147] The correlation degree between signal B and signal C = 0.78 * (1 / 0.31) * 0.1 ≈ 0.78 * 3.226 * 0.1 ≈ 0.252.

[0148] The correlation degree between signal B and signal D = 0.83 * (1 / 0.72) * 0.1 ≈ 0.83 * 1.389 * 0.1 ≈ 0.115.

[0149] The correlation degree of signal B and signal E = 0.56 * (1 / 1.35) * 0.1 ≈ 0.56 * 0.741 * 0.1 ≈ 0.041.

[0150] Therefore, the constructed correlation degree matrix is:

[0151]

[0152] In step 307, the correlation degree between the kth frame and the k+1th frame that meets the preset correlation condition is determined based on the correlation degree matrix. Each correlation signal pair includes a first signal and a second signal.

[0153] Further, a correlation degree threshold is preset, which is determined according to a large amount of correlation data of the unmanned aerial vehicle target signals and the non-target signals, and is used to distinguish effective correlation and ineffective correlation. Therefore, the unmanned aerial vehicle early warning system traverses each element in the correlation degree matrix, and when the correlation degree value of an element is greater than or equal to the preset correlation degree threshold, the suspected unmanned aerial vehicle target signal in the kth frame corresponding to the element and the suspected unmanned aerial vehicle target signal in the k+1th frame form a correlation signal pair, wherein the signal in the kth frame is the first signal, and the signal in the k+1th frame is the second signal.

[0154] In an embodiment, the preset correlation degree threshold is 0.2. The unmanned aerial vehicle early warning system traverses the correlation degree matrix constructed in step 306:

[0155] The correlation degree of element (1, 1) is 0.308 ≥ 0.2, so signal A (the first signal) and signal C (the second signal) form a correlation signal pair (A, C).

[0156] The correlation degree of element (1, 2) is 0.076 < 0.2, and no correlation signal pair is formed.

[0157] The correlation degree of element (1, 3) is 0.028 < 0.2, and no correlation signal pair is formed.

[0158] The correlation degree of element (2, 1) is 0.252 ≥ 0.2, so signal B (the first signal) and signal C (the second signal) form a correlation signal pair (B, C).

[0159] The correlation degree of element (2, 2) is 0.115 < 0.2, and no correlation signal pair is formed.

[0160] The correlation degree of element (2, 3) is 0.041 < 0.2, and no correlation signal pair is formed.

[0161] The finally determined correlation signal pairs are (A, C) and (B, C).

[0162] Step 308, based on the first spatial position difference value and the first signal feature similarity of the first signal and the second spatial position difference value and the second signal feature similarity of the second signal, determine the synergy coefficient of each associated signal pair.

[0163] Further, for each associated signal pair obtained in step 307, the unmanned aerial vehicle early warning system calculates the synergy coefficient according to the first spatial position difference value, the first signal feature similarity of the first signal, and the second spatial position difference value, the second signal feature similarity of the second signal. The synergy coefficient reflects the degree of cooperative matching of the two signals in the associated signal pair in spatial position and feature. Optionally, the embodiment of the application obtains the synergy coefficient by multiplying the signal feature similarity of the two signals and then combining the reciprocal of the spatial position difference value for weighting processing. The higher the synergy coefficient, the better the synergy of the associated signal pair, and the greater the possibility of belonging to the same target continuous signal.

[0164] Continuing the above embodiment, in the associated signal pair (A, C), the first spatial position difference value of signal A (the first signal) is 0.263 meters, and the first signal feature similarity is 0.81; the second spatial position difference value of signal C (the second signal) is 0.31 meters (assuming the position difference value of signal C with a signal in the k-1 frame), and the second signal feature similarity is 0.78. In the associated signal pair (B, C), the first spatial position difference value of signal B (the first signal) is 0.31 meters, and the first signal feature similarity is 0.78; the second spatial position difference value of signal C (the second signal) is 0.31 meters, and the second signal feature similarity is 0.78. The unmanned aerial vehicle early warning system calculates the synergy coefficient using the formula: Synergy coefficient = (first signal feature similarity * second signal feature similarity) * (1 / (first spatial position difference value + second spatial position difference value)) to obtain: the synergy coefficient of the associated signal pair (A, C) ≈ 1.093, and the synergy coefficient of the associated signal pair (B, C) ≈ 0.981.

[0165] Step 309, based on the synergy coefficient of each associated signal pair between the kth frame and the k+1 frame, determine the target signal sequence.

[0166] Further, the unmanned aerial vehicle early warning system determines the target signal sequence according to the synergy coefficient of each associated signal pair between the kth frame and the k+1 frame, specifically as the process of step 3091 to step 3094.

[0167] The embodiment of the present application quantifies the correlation degree between signals by constructing a correlation matrix, screens out the correlation signal pairs meeting the condition, and then calculates the synergy coefficient to determine the optimal correlation, and finally forms the target signal sequence, so that the false correlation can be effectively excluded, and the suspected signals belonging to the same unmanned aerial vehicle target in the continuous frames can be accurately correlated to form a coherent target signal sequence, which provides reliable basic data for subsequent analysis of the motion speed, direction, and continuity and regularity of the unmanned aerial vehicle target, and improves the accuracy and stability of the unmanned aerial vehicle early warning target tracking.

[0168] In an embodiment, the process of step 3091 to step 3094 includes:

[0169] Step 3091, based on the synergy coefficient of each correlation signal pair between the kth frame and the k+1th frame, the target correlation signal pairs meeting the preset synergy condition are combined to obtain the initial signal sequence segment between the kth frame and the k+1th frame.

[0170] Optionally, the unmanned aerial vehicle early warning system screens out the target correlation signal pairs meeting the preset synergy condition from the correlation signal pairs between the kth frame and the k+1th frame, and combines them into the initial signal sequence segment, so the synergy coefficient threshold is preset as the preset synergy condition, and the synergy coefficient threshold is determined according to the typical synergy characteristics of the unmanned aerial vehicle target signal. Further, the unmanned aerial vehicle early warning system checks the synergy coefficient of each correlation signal pair one by one, and when the synergy coefficient of a certain correlation signal pair is greater than or equal to the preset synergy coefficient threshold, the correlation signal pair is determined as the target correlation signal pair.

[0171] Further, the target correlation signal pairs are combined according to their order in the frame to form the initial signal sequence segment between the kth frame and the k+1th frame, and each initial signal sequence segment is composed of continuous target correlation signal pairs, reflecting the preliminary correlation relationship of the suspected unmanned aerial vehicle target signal between the two frames.

[0172] In an embodiment, the preset synergy coefficient threshold is 1. The correlation signal pairs between the kth frame and the k+1th frame and their synergy coefficients are as follows: the synergy coefficient of the correlation signal pair (A, C) is 1.093, the synergy coefficient of the correlation signal pair (B, D) is 0.95, and the synergy coefficient of the correlation signal pair (E, F) is 1.12.

[0173] Check the synergy coefficient of each correlation signal pair: the synergy coefficient of the correlation signal pair (A, C) is 1.093>1, which meets the preset synergy condition and is determined as the target correlation signal pair.

[0174] The synergy coefficient of the associated signal pair (B, D) is 0.95 < 1, which does not meet the preset synergy condition, and is not included in the target associated signal pair. The synergy coefficient of the associated signal pair (E, F) is 1.12 > 1, which meets the preset synergy condition, and is determined as the target associated signal pair. The target associated signal pairs are combined into an initial signal sequence segment, and two initial signal sequence segments [A→C] and [E→F] are obtained.

[0175] In step 3092, the spatial continuity index of each initial signal sequence segment is determined based on the spatial position difference of each signal in the target associated signal pair.

[0176] Further, for each initial signal sequence segment, the spatial continuity index of the segment is calculated by the UAV early warning system according to the spatial position difference of each signal in the target associated signal pair. The spatial continuity index is used to measure the continuity degree of the suspected UAV target signal in the spatial position in the initial signal sequence segment. The smaller the difference is, the more continuous the spatial position change is. In the embodiment of the present application, the mean value of the spatial position difference of all target associated signal pairs in the initial signal sequence segment is calculated, and the reciprocal of the mean value is taken as the spatial continuity index. The greater the spatial continuity index is, the more continuous and stable the spatial position change of the signals in the initial signal sequence segment is.

[0177] Continuing the above embodiment, for the initial signal sequence segment [A→C], the spatial position difference of the target associated signal pair (A, C) is 0.263 meters, the mean value of the spatial position difference is calculated as 0.263 meters, and therefore the spatial continuity index = 1 / mean value = 1 / 0.263 ≈ 3.802. For the initial signal sequence segment [E→F], the spatial position difference of the target associated signal pair (E, F) is 0.31 meters, the mean value of the spatial position difference is calculated as 0.31 meters, and therefore the spatial continuity index = 1 / 0.31 ≈ 3.226.

[0178] In step 3093, the feature consistency index of each initial signal sequence segment is determined based on the signal feature similarity of each signal in the target associated signal pair.

[0179] Further, the feature consistency index of each initial signal sequence segment is determined by the UAV early warning system according to the signal feature similarity of each signal in the target associated signal pair. The feature consistency index is used to measure the consistency degree of the suspected UAV target signal in the feature in the initial signal sequence segment. The higher the similarity is, the more consistent the feature is. In the embodiment of the present application, the mean value of the signal feature similarity of all target associated signal pairs in the initial signal sequence segment is calculated, and the mean value is the feature consistency index. The greater the feature consistency index is, the smaller the feature change of the signals in the initial signal sequence segment is, and the higher the consistency is.

[0180] Continuing the above example, the signal feature similarity of the target associated signal pair (A, C) for the initial signal sequence segment [A→C] is 0.81, thus the feature consistency index = 0.81. The signal feature similarity of the target associated signal pair (E, F) for the initial signal sequence segment [E→F] is 0.85, thus the feature consistency index = 0.85.

[0181] Step 3094, screening each initial signal sequence segment based on the spatial continuity index and the feature consistency index to obtain candidate signal sequence segments, and cross-frame splicing the candidate signal sequence segments to obtain the target signal sequence.

[0182] Further, the spatial continuity index threshold and the feature consistency index threshold are preset, and the initial signal sequence segment with the spatial continuity index greater than or equal to the spatial continuity index threshold and the feature consistency index greater than or equal to the feature consistency index threshold is determined as the candidate signal sequence segment by the unmanned aerial vehicle early warning system.

[0183] Further, the unmanned aerial vehicle early warning system cross-frame splices the candidate signal sequence segments, that is, when the end signal of a certain candidate signal sequence segment and the start signal of another candidate signal sequence segment belong to the same suspected unmanned aerial vehicle target signal in the continuous frames, the two candidate signal sequence segments are connected to form a longer signal sequence. In this way, the target signal sequence reflecting the continuous change of the suspected unmanned aerial vehicle target signal in multiple frames is gradually spliced.

[0184] Continuing the above example, the preset spatial continuity index threshold is 3, and the feature consistency index threshold is 0.8. The initial signal sequence segments are screened:

[0185] The spatial continuity index of the initial signal sequence segment [A→C] is 3.802>3, and the feature consistency index is 0.81>0.8, so it is determined as a candidate signal sequence segment.

[0186] The spatial continuity index of the initial signal sequence segment [E→F] is 3.226>3, and the feature consistency index is 0.85>0.8, so it is determined as a candidate signal sequence segment.

[0187] Suppose in the previous frame, there is a candidate signal sequence segment [G→A] with the end signal A, which is the same as the start signal A of the candidate signal sequence segment [A→C], so the candidate signal sequence segment [G→A] and the candidate signal sequence segment [A→C] are spliced to obtain [G→A→C].

[0188] There is a candidate signal sequence segment [H→E] with the end signal E, which is the same as the start signal E of the candidate signal sequence segment [E→F], so [H→E→F] is spliced.

[0189] The final target signal sequence is [G→A→C] and [H→E→F].

[0190] The embodiment of the application forms an initial signal sequence segment by screening qualified target associated signal pairs, screens candidate segments in combination with spatial continuity and feature consistency indexes, and finally obtains the target signal sequence by cross-frame splicing, so that the signal segments with spatial discontinuity and inconsistent features can be effectively eliminated, the target signal sequence formed can truly reflect the change of the suspected unmanned aerial vehicle target signal in continuous multiple frames, continuous and reliable signal sequence data for subsequent analysis of the motion characteristics of the suspected unmanned aerial vehicle target is provided, and the accuracy and effectiveness of the unmanned aerial vehicle early warning target tracking are improved.

[0191] In an embodiment, the process of steps 401 to 403 includes:

[0192] Step 401, based on the time interval and the spatial position change between adjacent frame signals in the target signal sequence, determine the instantaneous velocity component of each coordinate axis in the spatial rectangular coordinate system.

[0193] Optionally, the unmanned aerial vehicle early warning system obtains the time interval of the adjacent two frame signals in the target signal sequence, which is determined by the scanning period of the radar system (for example, the scanning interval of each frame is 1 second). Then, the three-dimensional coordinates (X k , Y k , Z k ) and (X k+1 , Y k+1 , Z k+1 ) of the adjacent two frame signals in the spatial rectangular coordinate system are extracted, and the spatial position change (ΔX=X k+1 -X k , ΔY=Y k+1 -Y k , ΔZ=Z k+1 -Z k ) on each coordinate axis is calculated.

[0194] Further, the unmanned aerial vehicle early warning system divides the position change of each coordinate axis by the time interval to obtain the instantaneous velocity component (v x , v y , v z ) on the X-axis, Y-axis and Z-axis, reflecting the instantaneous motion speed of the suspected unmanned aerial vehicle target signal in three dimensions.

[0195] Continuing the above embodiment, the target signal sequence is [G→A→C], wherein:

[0196] The coordinates of the k-th frame signal G are (90, 45, 25), the coordinates of the k+1-th frame signal A are (100, 50, 30), and the time interval between the two frames is Δt = 1 second. The spatial position change amount is calculated as follows: ΔX = 100-90 = 10 meters, ΔY = 50-45 = 5 meters, and ΔZ = 30-25 = 5 meters. The instantaneous velocity components are as follows: v x = ΔX / Δt = 10 / 1 = 10 meters / second, v y = ΔY / Δt = 5 / 1 = 5 meters / second, and v z = ΔZ / Δt = 5 / 1 = 5 meters / second.

[0197] Another set of adjacent frames: the k+1-th frame signal A (100, 50, 30) and the k+2-th frame signal C (110, 55, 35), and the time interval between the two frames is Δt = 1 second. The position change amount is as follows: ΔX = 10 meters, ΔY = 5 meters, and ΔZ = 5 meters. The instantaneous velocity components are as follows: v x = 10 meters / second, v y = 5 meters / second, and v z = 5 meters / second.

[0198] In step 402, the instantaneous resultant velocity of the adjacent frame signals and the motion direction angle component of the adjacent frame signals in the spatial rectangular coordinate system are determined based on the instantaneous velocity components of the adjacent frame signals.

[0199] Further, the instantaneous resultant velocity v of the suspected unmanned aerial vehicle target signal in space is obtained by taking the square root of the sum of the three-dimensional velocity components, which reflects the actual motion speed of the suspected unmanned aerial vehicle target signal in space.

[0200] Further, the motion direction angle component includes the azimuth angle (the included angle with the X-axis in the XY plane) and the pitch angle (the included angle with the XY plane). The azimuth angle θ is calculated by the arctangent function: θ = arctan(v y / v x ; and the pitch angle φ is calculated by the arcsine function: φ = arcsin(v z / v). The two angles together determine the motion direction of the suspected unmanned aerial vehicle target signal in space.

[0201] Continuing with the example of the adjacent frames from signal G to signal A: the instantaneous velocity components are as follows: v x = 10 meters / second, v y = 5 meters / second, and v z= 5 m / s, so the instantaneous resultant velocity v ~ 12.25 m / s. The azimuth angle Θ = arctan(5 / 10) ~ 26.565° (the angle with the positive direction of the X-axis in the XY plane). The pitch angle φ = arcsin(5 / 12.25) ~ 24.1° (the angle with the XY plane). The results of the adjacent frame calculation of the signals A to C are the same: the instantaneous resultant velocity ~ 12.25 m / s, the azimuth angle ~ 26.565°, and the pitch angle ~ 24.1°.

[0202] In step 403, the instantaneous resultant velocities of the adjacent frame signals are arranged in sequence according to the frames to generate the movement velocities of the suspected UAV target signal in the continuous frames, and the direction angle components of the adjacent frame signals in the spatial rectangular coordinate system are combined to generate the movement direction of the suspected UAV target signal in the continuous frames.

[0203] Further, the UAV early warning system arranges the instantaneous resultant velocities of each adjacent frame in sequence according to the frames (such as the k→k+1 frame, the k+1→k+2 frame, and so on), forms a velocity sequence, and the velocity sequence is the movement velocity of the suspected UAV target signal in the continuous frames, reflecting the change of the velocity over time.

[0204] For the movement direction, the UAV early warning system combines the azimuth angle and the pitch angle components of the adjacent frames in sequence according to the frames to form a direction angle sequence, each direction angle is composed of the corresponding azimuth angle and pitch angle, and together constitutes the movement direction of the suspected UAV target signal in the continuous frames, reflecting the change of the spatial pointing over time.

[0205] Continue to take the example that the target signal sequence [G→A→C] contains two groups of adjacent frames:

[0206] The k→k+1 frame (G→A): the instantaneous resultant velocity is 12.25 m / s, the azimuth angle is 26.565°, and the pitch angle is 24.1°. The k+1→k+2 frame (A→C): the instantaneous resultant velocity is 12.25 m / s, the azimuth angle is 26.565°, and the pitch angle is 24.1°.

[0207] The movement velocity is arranged as: [12.25 m / s, 12.25 m / s]. The movement direction is combined as: [(26.565°, 24.1°), (26.565°, 24.1°)], indicating a stable direction in the continuous frames.

[0208] The embodiment of the present application completely constructs the motion speed and the motion direction of the suspected unmanned aerial vehicle target signal by decomposing the three-dimensional velocity component, synthesizing the instantaneous resultant velocity and the direction angle, and integrating the continuous frame data. Therefore, not only the motion speed of the target in the three-dimensional space can be accurately quantified, but also the motion direction of the target can be determined, and the change characteristics of the speed and the direction over time are retained. Subsequent analysis of the continuity (such as whether the speed continuously exists) and the regularity (such as whether the speed and the direction are stable) of the motion provides accurate and comprehensive kinematic data, which ensures that whether the target is an unmanned aerial vehicle can be determined based on objective motion characteristics, and the reliability of the early warning logic is improved.

[0209] In an embodiment, the processes of steps 404 to 408 include:

[0210] Step 404, determining the change rate of the instantaneous resultant velocity between the continuous frames based on the motion speed of the continuous frames, and determining the direction change amount between the continuous frames based on the motion direction of the continuous frames.

[0211] Optionally, the unmanned aerial vehicle early warning system processes the motion speed: for the motion speed sequence of the continuous frames (i.e., the instantaneous resultant velocities of each adjacent frame), the difference between the two adjacent instantaneous resultant velocities and the ratio of the corresponding time interval are calculated to obtain the change rate of the instantaneous resultant velocity, which reflects the speed change.

[0212] Further, the unmanned aerial vehicle early warning system processes the motion direction: for the motion direction sequence of the continuous frames (i.e., the direction angle components of each adjacent frame), the included angle between the two adjacent direction angle components is calculated by the space vector included angle formula to obtain the direction change amount, which reflects the direction change.

[0213] Continuing the above embodiment, the motion speed of the target signal sequence [G→A→C] is [12.25 m / s, 12.25 m / s], and the time interval of the adjacent frames is 1 second. The change rate of the instantaneous resultant velocity: the change rate of the k+1→k+2 frame = (12.25-12.25) / 1 = 0 m / s2.

[0214] The motion direction is [(26.565°, 24.1°), (26.565°, 24.1°)], so the direction change amount: calculated by the space vector included angle formula, the included angle between the two direction vectors is 0°, i.e., the direction change amount = 0°.

[0215] Step 405, determining the speed continuity index between the continuous frames based on the change rate of the instantaneous resultant velocity between the continuous frames, and determining the direction continuity index between the continuous frames based on the direction change amount between the continuous frames.

[0216] Further, the unmanned aerial vehicle early warning system takes the absolute value of the change rate of the instantaneous resultant velocity between the continuous frames as the speed continuity index. The smaller the index is, the slower the speed change is, and the better the continuity is.

[0217] Further, the UAV early warning system takes the direction change amount between consecutive frames as the direction persistence index directly. The smaller the index is, the more gentle the direction change is, and the better the persistence is.

[0218] Continuing the above example, the speed persistence index = 0 m / s2. The direction persistence index = 0°.

[0219] Step 406, based on the motion speed of consecutive frames and the average speed in the target signal sequence, determine the speed regularity deviation amount between consecutive frames, and based on the motion direction of consecutive frames and the relative direction in the target signal sequence, determine the direction regularity deviation amount between consecutive frames.

[0220] Further, the UAV early warning system calculates the average of all instantaneous resultant velocities in the target signal sequence (average speed), and then calculates the absolute difference between the motion speed of each consecutive frame and the average speed as the speed regularity deviation amount. The smaller the deviation amount is, the more the speed conforms to the overall trend.

[0221] Further, for the motion direction, first determine the relative direction of each direction in the target signal sequence relative to the initial direction (such as the cumulative change of the direction angle), and then calculate the absolute difference between the motion direction of each consecutive frame and the relative direction as the direction regularity deviation amount. The smaller the deviation amount is, the more the direction conforms to the overall trend.

[0222] Continuing the above example, the motion speed of the target signal sequence [G→A→C] is [12.25 m / s, 12.25 m / s], and the average speed = (12.25+12.25) / 2 = 12.25 m / s. Therefore, the speed regularity deviation amount: the deviation amount of the k→k+1 frame = |12.25-12.25| = 0 m / s; the deviation amount of the k+1→k+2 frame = 0 m / s.

[0223] The motion direction has no change relative to the initial direction, and the relative direction is (26.565°, 24.1°). Therefore, the direction regularity deviation amount: both frames are |26.565°-26.565°| = 0°, |24.1°-24.1°| = 0°, and the comprehensive direction regularity deviation amount = 0°.

[0224] Step 407, if the speed persistence index of consecutive frames is less than or equal to the preset speed index threshold, and the direction persistence index between consecutive frames is less than or equal to the preset direction index threshold, it is determined that the motion has persistence.

[0225] Further, preset speed index threshold (such as 1 m / s2) and direction index threshold (such as 5°) are set. If the speed persistence index between the continuous frames is all less than or equal to the preset speed index threshold, and the direction persistence index is all less than or equal to the preset direction index threshold, it is considered that the speed and direction change gently, and there is no sudden interruption or dramatic change, and the UAV early warning system determines that the motion has persistence, otherwise, it is determined that the motion does not have persistence.

[0226] In an embodiment, since the speed persistence index is 0 m / s2≤1 m / s2, and the direction persistence index is 0°≤5°, the UAV early warning system determines that the motion has persistence.

[0227] In step 408, if the speed regularity deviation amount of the continuous frames is less than or equal to the speed deviation threshold, and the direction regularity deviation amount between the continuous frames is less than or equal to the preset direction deviation threshold, it is determined that the motion has regularity.

[0228] Further, preset speed deviation threshold (such as 0.5 m / s) and direction deviation threshold (such as 3°) are set. If the speed regularity deviation amount between the continuous frames is all less than or equal to the speed deviation threshold, and the direction regularity deviation amount is all less than or equal to the preset direction deviation threshold, it is considered that the speed and direction fluctuate around the overall trend with small amplitude, and there is no irregular mutation, and the UAV early warning system determines that the motion has regularity, otherwise, it is determined that the motion does not have regularity.

[0229] In an embodiment, since the speed regularity deviation amount is 0 m / s<0.5 m / s, and the direction regularity deviation amount is 0°<3°, the UAV early warning system determines that the motion has regularity.

[0230] The embodiment of the present application can accurately determine whether the motion of the suspected UAV target has persistence and regularity by quantifying the change rate and deviation amount of the speed and direction, and combining the preset threshold, so that the verification is performed from two dimensions of dynamic change (persistence) and overall trend consistency (regularity), the signals with sudden speed change, sudden direction change or irregular fluctuation (such as flying birds and clutter) are excluded, the target meeting the UAV motion characteristics is accurately locked, and a decisive basis is provided for determining the UAV target and generating the early warning information, so that the target recognition accuracy and early warning reliability of the UAV early warning are improved.

[0231] Optionally, referring to Figure 2 , Figure 2 is a flowchart of the UAV radar early warning system provided by the present application, and the UAV radar early warning system comprises:

[0232] The signal acquisition and extraction module 210 is configured to acquire the radar echo signal of the airspace based on the radar system, and perform feature extraction on the radar echo signal to obtain the time domain feature and the frequency domain feature of each signal in the radar echo signal.

[0233] The signal screening module 220 is configured to screen signals based on the feature screening condition in combination with the time domain feature and the frequency domain feature of each signal to determine the suspected unmanned aerial vehicle target signal in the radar echo signal.

[0234] The signal correlation module 230 is configured to perform signal correlation on the suspected unmanned aerial vehicle target signals in continuous multiple frames to determine the spatial position difference and the signal feature similarity of the suspected unmanned aerial vehicle target signals in adjacent frames, and determine the target signal sequence based on the spatial position difference and the signal feature similarity.

[0235] The motion trajectory analysis module 240 is configured to determine the motion speed and the motion direction of the suspected unmanned aerial vehicle target signal in continuous frames based on the target signal sequence, and determine whether the motion of the suspected unmanned aerial vehicle target has continuity and regularity based on the motion speed and the motion direction.

[0236] The early warning determination output module 250 is configured to determine that there is an unmanned aerial vehicle target if the motion has continuity and regularity, generate early warning information and output.

[0237] The embodiment of the present application avoids missing the target due to a single signal strength indicator by extracting multiple features to screen the suspected unmanned aerial vehicle target signal, and further confirms the authenticity of the suspected unmanned aerial vehicle target signal through multi-frame signal correlation and motion trajectory analysis. Even if the signal strength fluctuates, as long as the features and the motion conform to the regularity, the suspected unmanned aerial vehicle target signal can be detected, thereby reducing the false alarm. Furthermore, through multi-frame correlation, the spatial position and the feature are required to be similar, which excludes transient interference signals. The motion trajectory analysis judges the continuity and regularity of the motion, which excludes irregular interference signals, thereby reducing the false alarm and improving the reliability of the unmanned aerial vehicle radar early warning.

[0238] Please refer to Figure 3 , Figure 3 An embodiment of the electronic device provided by the embodiment of the present application is shown in the figure. As shown in the figure, Figure 3 The embodiment of the present application provides an electronic device 300, which includes a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, the following steps are implemented:

[0239] Based on the radar system, radar echo signals in the airspace are collected, and feature extraction is performed on the radar echo signals to obtain the time domain feature and the frequency domain feature of each signal in the radar echo signals;

[0240] Based on the feature screening condition in combination with the time domain feature and the frequency domain feature of each signal, signal screening is performed to determine the suspected unmanned aerial vehicle target signal in the radar echo signal;

[0241] The signal correlation is performed on the suspected unmanned aerial vehicle target signals of continuous multiple frames, the spatial position difference and signal feature similarity of the suspected unmanned aerial vehicle target signals in adjacent frames are determined, and based on the spatial position difference and the signal feature similarity, the target signal sequence is determined;

[0242] Based on the target signal sequence, the motion speed and motion direction of the suspected unmanned aerial vehicle target signal in the continuous frames are determined, and based on the motion speed and the motion direction, whether the motion of the suspected unmanned aerial vehicle target has continuity and regularity is determined.

[0243] If the motion has continuity and regularity, it is determined that there is an unmanned aerial vehicle target, and an early warning information is generated and output.

[0244] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0245] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0246] The present application is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The functions specified in one flow or multiple flows and / or blocks

[0247] These computer program instructions can also be stored in a computer-readable storage medium that can guide the computer or other programmable data processing devices to work in a specific way, so that the instructions stored in the computer-readable storage medium produce a product including instruction devices that implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The functions specified in one flow or multiple flows and / or blocks

[0248] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 These computer program instructions can also be loaded into a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 These computer program instructions can also be loaded into a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks.

[0249] While the preferred embodiments of the application have been described, it should be apparent that a little thought and experimentation can lead to the development of other techniques and approaches that are widely equivalent to those described above. Accordingly, no limitation is intended to the scope of the protection granted to the application that can encompass all techniques and approaches comparable to those described and falling within the scope of the claims below and the scope of equivalents thereof.

[0250] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A method for unmanned aerial vehicle radar warning, characterized in that, The method comprises: collecting radar echo signals of an airspace based on a radar system, and performing feature extraction on the radar echo signals to obtain time domain features and frequency domain features of each signal in the radar echo signals; performing signal screening based on feature screening conditions in combination with the time domain features and the frequency domain features of each signal to determine suspected unmanned aerial vehicle target signals in the radar echo signals; performing signal correlation on the suspected unmanned aerial vehicle target signals of continuous multiple frames to determine spatial position difference and signal feature similarity of the suspected unmanned aerial vehicle target signals in adjacent frames, and determining a target signal sequence based on the spatial position difference and the signal feature similarity; determining a motion speed and a motion direction of the suspected unmanned aerial vehicle target signals in continuous frames based on the target signal sequence, and determining whether the motion of the suspected unmanned aerial vehicle target has continuity and regularity based on the motion speed and the motion direction; if the motion has continuity and regularity, it is determined that there is an unmanned aerial vehicle target, and a warning information is generated and output.

2. The UAV radar warning method of claim 1, wherein, The method of determining the motion speed and the motion direction of the suspected unmanned aerial vehicle target signals in continuous frames based on the target signal sequence comprises: determining each instantaneous speed component of adjacent frame signals in a three-dimensional coordinate system of a spatial rectangular coordinate system based on a time interval and a change amount of a spatial position between the adjacent frame signals in the target signal sequence; determining an instantaneous resultant speed of the adjacent frame signals and a motion direction angle component of the adjacent frame signals in the spatial rectangular coordinate system based on each instantaneous speed component of the adjacent frame signals; arranging the instantaneous resultant speeds of the adjacent frame signals in sequence to generate a motion speed of the suspected unmanned aerial vehicle target signals in continuous frames, and combining the motion direction angle components of the adjacent frame signals in the spatial rectangular coordinate system to generate a motion direction of the suspected unmanned aerial vehicle target signals in continuous frames. 3.The unmanned aerial vehicle radar warning method of claim 1, wherein, The method of determining whether the motion of the suspected unmanned aerial vehicle target has continuity and regularity based on the motion speed and the motion direction comprises: determining a change rate of the instantaneous resultant speed between continuous frames based on the motion speed of the continuous frames, and determining a direction change amount between the continuous frames based on the motion direction of the continuous frames; determining a speed continuity index between the continuous frames based on the change rate of the instantaneous resultant speed between the continuous frames, and determining a direction continuity index between the continuous frames based on the direction change amount between the continuous frames; determining a speed regularity deviation amount between the continuous frames based on the motion speed of the continuous frames and a speed average in the target signal sequence, and determining a direction regularity deviation amount between the continuous frames based on the motion direction of the continuous frames and a relative direction in the target signal sequence; if the speed continuity index between the continuous frames is less than or equal to a preset speed index threshold, and the direction continuity index between the continuous frames is less than or equal to a preset direction index threshold, it is determined that the motion has continuity; if the speed regularity deviation amount between the continuous frames is less than or equal to a speed deviation threshold, and the direction regularity deviation amount between the continuous frames is less than or equal to a preset direction deviation threshold, it is determined that the motion has regularity. 4.The unmanned aerial vehicle radar warning method of claim 1, wherein, The signal correlation is performed on the suspected unmanned aerial vehicle target signals in the continuous multiple frames, spatial position difference values and signal feature similarities of the suspected unmanned aerial vehicle target signals in adjacent frames are determined, and the method comprises the following steps: A space rectangular coordinate system is established with the phase center of the radar antenna as the origin; the horizontal axis of the space rectangular coordinate system points to the front of the radar, the vertical axis is horizontally to the right, and the vertical axis is vertically upward; A feature vector is constructed based on the spatial position parameters and signal feature parameters of each suspected unmanned aerial vehicle target signal in the space rectangular coordinate system; the signal feature parameters comprise time domain features and frequency domain features; For the kth suspected unmanned aerial vehicle target signal, the position difference value basis of the suspected unmanned aerial vehicle target signal in the kth frame is determined according to the first spatial position parameter in the feature vector thereof in combination with the second spatial position parameter in the feature vector of all suspected unmanned aerial vehicle target signals in the k+1th frame; The feature similarity basis of the suspected unmanned aerial vehicle target signal in the kth frame is determined according to the first signal feature parameter in the feature vector thereof in combination with the second signal feature parameter in the feature vector of all suspected unmanned aerial vehicle target signals in the k+1th frame; The spatial position difference values and signal feature similarities of the suspected unmanned aerial vehicle target signals in adjacent frames are determined based on the position difference value basis and the feature similarity basis of the suspected unmanned aerial vehicle target signal in the kth frame.

5. The UAV radar warning method of claim 4, wherein, The spatial position difference values and signal feature similarities of the suspected unmanned aerial vehicle target signals in adjacent frames are determined based on the position difference value basis and the feature similarity basis of the suspected unmanned aerial vehicle target signal in the kth frame, and the method comprises the following steps: The first distribution feature coefficients of all suspected unmanned aerial vehicle target signals in the k+1th frame are determined based on the position difference value basis of the suspected unmanned aerial vehicle target signal in the kth frame; The second distribution feature coefficients of all suspected unmanned aerial vehicle target signals in the k+1th frame are determined based on the feature similarity basis of the suspected unmanned aerial vehicle target signal in the kth frame; The spatial position difference values of the suspected unmanned aerial vehicle target signals in adjacent frames are determined based on the position difference value basis of the suspected unmanned aerial vehicle target signal in the kth frame and the first distribution feature coefficients; The signal feature similarities of the suspected unmanned aerial vehicle target signals in adjacent frames are determined based on the feature similarity basis of the suspected unmanned aerial vehicle target signal in the kth frame and the second distribution feature coefficients.

6. The UAV radar warning method of claim 5, wherein, The target signal sequence is determined based on the spatial position difference values and the signal feature similarities, and the method comprises the following steps: An association degree matrix is constructed based on the spatial position difference values and the signal feature similarities of the suspected unmanned aerial vehicle target signals in adjacent frames; the matrix elements in the association degree matrix represent the association degrees between the suspected unmanned aerial vehicle target signals in the kth frame and the k+1th frame; The association signal pairs whose association degrees between the kth frame and the k+1th frame satisfy a preset association condition are determined based on the association degree matrix; each association signal pair comprises a first signal and a second signal; The coordination coefficients of each association signal pair are determined based on the first spatial position difference value and the first signal feature similarity of the first signal and the second spatial position difference value and the second signal feature similarity of the second signal; The target signal sequence is determined based on the coordination coefficients of each association signal pair between the kth frame and the k+1th frame.

7. The UAV radar warning method of claim 6, wherein, The target signal sequence is determined based on the correlation coefficient of each associated signal pair between the kth frame and the k+1th frame. The target associated signal pairs meeting the preset correlation condition are combined based on the correlation coefficient of each associated signal pair between the kth frame and the k+1th frame, to obtain an initial signal sequence segment between the kth frame and the k+1th frame. The spatial continuity index of each initial signal sequence segment is determined based on the spatial position difference of each signal in the target associated signal pair. The feature consistency index of each initial signal sequence segment is determined based on the signal feature similarity of each signal in the target associated signal pair. The candidate signal sequence segment is obtained by screening each initial signal sequence segment based on the spatial continuity index and the feature consistency index, and the target signal sequence is obtained by cross-frame splicing the candidate signal sequence segment.

8. An unmanned aerial vehicle radar warning system characterized by, The signal acquisition and extraction module is configured to acquire radar echo signals in a space domain based on a radar system, and extract features of the radar echo signals to obtain time domain features and frequency domain features of each signal in the radar echo signals. The signal screening module is configured to screen signals based on feature screening conditions in combination with the time domain features and the frequency domain features of each signal, to determine suspected unmanned aerial vehicle target signals in the radar echo signals. The signal correlation module is configured to correlate the suspected unmanned aerial vehicle target signals in consecutive multiple frames, to determine spatial position differences and signal feature similarities of the suspected unmanned aerial vehicle target signals in adjacent frames, and to determine a target signal sequence based on the spatial position differences and the signal feature similarities. The motion trajectory analysis module is configured to determine a motion speed and a motion direction of the suspected unmanned aerial vehicle target signals in consecutive frames based on the target signal sequence, and to determine whether the motion of the suspected unmanned aerial vehicle target signals has continuity and regularity based on the motion speed and the motion direction. The early warning determination and output module is configured to determine that there is an unmanned aerial vehicle target if the motion has continuity and regularity, to generate early warning information, and to output the early warning information. The electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, the method for early warning of an unmanned aerial vehicle radar according to any one of claims 1 to 7 is implemented.

9. An electronic device, comprising: The processor executes the program to implement the method for early warning of an unmanned aerial vehicle radar according to any one of claims 1 to 7.

10. A non-transitory computer readable storage medium having stored therein a computer software program, characterized in that, ​

Images